Bulletin of the Asia Institute

Evo ṣuyadi Essays in Honor of Richard Salomon’s 65th Birthday

New Series/Volume 23

2009

Edited by Carol Altman Bromberg, Timothy J. Lenz, and Jason Neelis

Published with the assistance of the Neil Kreitman Foundation (U.K.) Contents

Foreword 1 Michael Shapiro Richard Salomon: A Personal Tribute 5 Mark Allon A Gāndhārī Version of the Story of the Merchants Tapussa and Bhallika 9 Stefan Baums Inscribed Buddhist Tablets from Merv 21 Daniel Boucher What Do We Mean by “Early” in the Study of the Early Mahāyāna—and Should We Care? 33 Robert L. Brown Telling the Story in Art of the Monkey’s Gift of Honey to the Buddha 43 Collett Cox What’s in a Name? School Affiliation in an Early Buddhist Gāndhārī Manuscript 53 Harry Falk Making Wine in under Buddhist Monastic Supervision 65 Andrew Glass Bha 79 Paul Harrison Verses by Śāntideva in the Śikṣāsamuccaya: A New English Translation 87 Jens-Uwe Hartmann The Foolish Cat and the Clever Mouse: Another Parable from an Unknown Story Collection 105 Stephanie W. Jamison An Indo-Iranian Priestly Title Lurking in the Rig Veda? An Indic Equivalent to Avestan karapan 111 Seishi Karashima On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha 121 Klaus Karttunen Gandhāra and the Greeks 131 Timothy Lenz Ephemeral ; Magical Hope 135 Abdur Rehman A Note on the Etymology of Gandhāra 143 Juhyung Rhi The Garuḍa and the Nāgī/Nāga in the Headdresses of Gandhāran : Locating Textual Parallels 147 Ludo Rocher and Indian Epigraphy and the Asiatic Society: The First Rosane Rocher Fifty Years 159 Gregory Schopen Regional Languages and the Law in Some Early North Indian Buddhist Monasteries and Convents 171 Martin Schwartz Sartha- and Other Caravan Words 179 Jonathan A. Silk The Nature of the Verses of the Kāśyapaparivarta 181 Nicholas Sims-Williams Some Bactrian Inscriptions on Silver Vessels 191 Peter Skilling Prakrit Prajñāpāramitās: Northwest, South, and Center: Gleanings from Avalokitavrata and Haribhadra 199 Ingo Strauch Inscribed Objects from Greater Gandhāra 209 Michael Willis Avalokiteśvara of the Six Syllables: Locating the Practice of the “Great Vehicle” in the Landscape of Central India 221

Review jongeward, errington, salomon, and baums. Gandharan Buddhist Reliquaries (Jason Neelis) 231

Books Received 237 Abbreviations 239

v On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha*

Seishi Karashima international research institute for advanced buddhology, soka university

It is now widely recognised that quite a few of Sukhāvatīvyūha, namely the Da āmítuó jīng early Mahāyāna Buddhist scriptures, written in (大阿彌陀經, T. 12, no. 362), which was most (Buddhist) , were transmitted originally probably translated by Zhiloujiachen 支婁迦讖 in Middle Indic and subsequently “translated” or Lokakṣema (fl. ca 170–190 c.e.),3 this Bud- gradually into (Buddhist) Sanskrit. Such (Buddhist) dha is always described as possessing incompa- Sanskrit texts are, in other words, the result of rable light4 but not limitless life. He even enters constant Sanskritization, wrong back-formations, parinirvāṇa (309a15) so as to be succeeded by additions and interpolations over the centuries. *Olokitasvara (hélóugèn 廅樓亘 < Avalokita­ Chinese translations, on the other hand, particu- svara5). Therefore, he was never seen as having larly those which were made from the second to “limitless life” in this earliest version. the sixth century c.e. and thus antedating­ most In the present Sanskrit version of the Larger of the extant Sanskrit manuscripts, may provide Sukhāvatīvyūha, the form Amitābha occurs substantial clues to the origination and devel- only in the prose,6 while Amitāyu is found only opment of Buddhist scriptures.1 If we read early in the verses7—the same is true in the case of the Chinese translations carefully in the light of San- Sanskrit fragments of the same scripture discov- skrit, Tibetan and other Chinese versions, while ered in Afghanistan, which is now preserved in taking into consideration the newly discovered the Schøyen Collection.8 Although in the present old Gāndhārī fragments of various texts, we may Sanskrit manuscripts from Nepal dating from the be able to reconstruct earlier and more original twelfth century Amitāyus, the Sanskritized form features of early Mahāyāna Buddhist scriptures of Amitāyu, appears seven times in the prose, and trace their transmissions. most of which lack parallels in the Tibetan and (or) Chinese translations, their occurrences are therefore presumably either later interpolations I. Amitābha (“Limitless Light”) > or substitutions in place of Amitābha, which Amitāyus (“Limitless Life”) took place when Amitāyus became more popular than Amitābha.9 Concerning the two names of the Buddha in the Concerning why the form Amitāyu occurs land of Sukhāvatī, namely Amitābha (“Limit- only in the verses, my hypothesis is as follows. less Light”) and Amitāyus (“Limitless Life”), When Amitābha or rather its Middle Indic form I have suggested elsewhere2 that Amitābha Amitāha10 was used in the verses and the ending (“Limitless Light”; wúliàngguāng 無量光 in Chi- of its nominative singular forms Amitābho / MI nese) is a more original form and that only later Amitāho needed to be shortened for metrical rea- did it evolve gradually into Amitāyus (“Limit- sons, the name of the Buddha must have become less Life”; wúliàngshòu 無量壽 in Chinese). In Amitābhu / MI *Amitāhu.11 The Middle Indic the oldest Chinese translation of the Larger form *Amitāhu would have been pronounced as

121 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

*Amitā˘ ’u or *Amidā˘ ’u in Gāndhārī, in which the SP(KN) 455.2. Amitābha-(nāyaka~) (= K, C5, C6, h is silent. The Gāndhārī forms -ā˘ ’u could have R etc.) / L2, L3, Bj, N2, T8, B etc. Amitābhu- / been interpreted also as coming from Skt. āyus O, D2, L1, C4, N1, etc. Amitāyu- (= Tib. Kho. (“life”); cf. MI āu < Skt. āyus. Hence, the form ga 45b1. TSHe mtha yas pa, Tib. Kanj. TSHe *Amitā’u (or *Amidā’u) could have meant both mtha’ yas pa) “Limitless Light” and “Limitless Life.” Later, SP(KN) 455.5. Amitābha-(nāyaka~) (= P1, A2 however, when the trend to Sanskritize earlier etc.) / C4. Amitābhu- / L2, L3, Bj, C5, C6, Mahāyāna Buddhist —which must have B etc. tatha loka-(nāyaka~) / O, D2, K, N1. contained more Middle Indic forms than the Amitāyu- (= Tib. Kho. ga 45b2. TSe mtha’ yas present Sanskrit versions—emerged, presum- pa, Tib. Kanj. TSHe mtha’ yas pa) ably around the beginning of the third century c.e.,12 the form *Amitā’u (or *Amidā’u) in the The confusion among these three forms in verses was Sanskritized incorrectly as Amitāyu the verses of the Lotus support the pos- by somebody who probably thought the designa­ sibility that the development Amitābha > tion “Limitless Life” was a more appropriate epi­ Amitā(b)hu > Amitāyu took place in the Larger thet for this Buddha. Thus, the name Amitāyu / Sukhāvatīvyūha. Amitāyus (“Limitless Life”) was eventually cre- As we have seen above, in the Larger Su­ ated as a hyper-form.13 khāvatīvyūha, the form Amitābha occurs in This hypothesized development, namely the prose part, while Amitāyu is found only in Amitābha > Amitā(b)hu > Amitāyu in the Larger the verses. Contrary to this, in the Smaller Su­ Sukhāvatīvyūha, may be supported by instances khāvatīvyūha, which is composed solely in prose, of confusion among the three forms, which take the secondary and later formed name Amitāyus is place in verses 29, 30 and 32 in the Saman­ used exclusively except in one place where the tamukha-parivarta of the Lotus Sutra.14 At first, Buddha explains why he has two names: we shall quote the editio princeps, the edition by H. Kern and B. Nanjio. The verses in question tat kiṃ manyase Śāriputra! kena kāraṇena SP(KN) 454.5–455.5) read as follows: sa tathāgato ’mitāyur nāmôcyate? tasya khalu punaḥ Śāriputra! tathāgatasya teṣāṃ sthita dakṣiṇavāmatas tathā vījayanta ca manuṣyāṇām aparimitam āyuṣpramāṇaṃ. Amitābhanāyakam | tena kāraṇena sa tathāgato ’mitāyur māyopama te samādhinā sarvakṣetra (read nāmôcyate. tasya ca Śāriputra! tathāgatasya °tre) jina gandha (read gatvā) pūjiṣu || 29 || daśa kalpā anuttarāṃ samyaksaṃbodhim diśa paścima yatra sukhākarā lokadhātu abhisaṃbuddhasya. tat kiṃ manyase virajā Sukhāvatī | Śāriputra! kena kāraṇena sa tathāgato yatra eṣa Amitābhanāyakaḥ saṃprati tiṣṭhati ’mitābho nāmôcyate? tasya khalu punaḥ sattvasārathiḥ || 30 || Śāriputra! tathāgatasyâbhâpratihatā . . . sarvabuddhakṣetreṣu. tena kāraṇena sa so caiva Amitābhanāyakaḥ padmagarbhe tathāgato ’mitābho nāmôcyate. (Fujita 2011: ­viraje manorame | 87.18–88.8) siṃhāsani saṃniṣaṇṇako Śālarājo va yathā virājate || 32 || “What do you think, O Śāriputra? Why is that Tathāgata called Amitāyus? Now, O Śāriputra, In the above-quoted edition, the form Amitābha the length of that Tathāgata’s life and of those occurs three times. However, readings Amitābha, men there is immeasurable. Therefore, that Amitābhu and Amitāyu vary among manuscripts Tathāgata is called Amitāyus. And ten kalpas and the Tibetan translation of the same scripture: have passed, O Śāriputra, since this Tathāgata attained to unsurpassed, perfect, enlighten- SP(KN) 454.5. Amitābha-(nāyaka~) (= C5, ment. What do you think, O Śāriputra? Why C6, R etc.15) (= Tib. Kanj.16 sNang ba mtha’ is that Tathāgata called Amitābha? Now, O yas) / L2, L3, K, Bj, N2, B etc. Amitābhu- / Śāriputra, the light of this Tathāgata spreads O, D2, L1, C4, N1 etc. Amitāyu- (= Tib. unimpeded over all the Buddha-lands. There- Kho. ga 45a717. TSe mtha’ yas) fore, that Tathāgata is called Amitābha.”

122 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

Kōtatsu Fujita has assumed that this Buddha tion of the Larger Sukhāvatīvyūha, namely the was worshipped under different names by different Da āmítuó jīng might have been Amitāha or groups,18 and those who worshipped Amitābha *Amidāha (probably pronounced as *Amitā’a composed the Larger Sukhāvatīvyūha, while or *Amidā’a), which are Middle Indic forms of those who worshipped Amitāyus composed the Amitābha.24 Smaller Sukhāvatīvyūha.19 He also maintains Throughout the second oldest Chinese trans­ that these two sutras were composed almost at lation of the same scripture, namely the Wúliàng­ the same time by these two groups of worshippers qīngjìng Píngdĕngjué jīng 無量清淨平等覺經 (T. 12, of the same Buddha. This theory is, however, very no. 361) by Zhi Qian 支謙 (fl. ca. 220–257 c.e.), arbitrary. which is none other than a “modified version” of I presume that the name Amitāyu (= Amitāyus; the Da āmítuó jīng, the Buddha’s name is changed “Limitless Life”), which had originally been to wúliàngqīngjìng(fó) 無量清淨(佛) “(Buddha) of a hyper-form­ in the verses, originating from Infinite Purity.” Amitābha (“Limitless Light”), grew more and Jan Nattier assumes that this name was origi- more popular as it became widely accepted as nally a translation of the title of the text, namely a more appropriate epithet in its own right for Amitābha-vyūha.25 I agree with her assumption in this Buddha and finally came to be used in prose principle, but with some differences. As we shall as well. In this way, the same Buddha came to see later, in the older Sanskrit manuscripts, the have two different names, with different mean- title of the sutra reads Amitābha-vyūha instead ings. However, those who worshipped this Bud- of Sukhāvatī-vyūha. The original title of the older dha, probably knew that these two names referred Chinese translations and the Tibetan one seem to to one and the same Buddha and did not think it have been Amitābha-vyūha as well. The underly- strange, as it is common in many cultures that ing title of the oldest Chinese translations, namely one god has different names or various epithets. the Da āmítuó jīng and the Wúliàngqīngjìng From the transliterations used in the Da āmítuó Píngdĕngjué jīng, was probably its Middle Indic jīng, it is evident that the underlying Indian text form *Amitāha-(or Amidāha-)vyūha (< Amitābha- of this oldest Chinese translation of the Larger Su­ vyūha). Lokakṣema transliterated the Buddha’s khāvatīvyūha had been transmitted in Middle Indic, name as Āmítuó 阿彌陀 and, therefore, translated most probably Gāndhārī. In contrast, the Smaller the title of the sutra as Āmítuó jīng 阿彌陀經–later Sukhāvatī­ vyūha­ , which appeared later than the it was altered to Da āmítuó jīng.26 In Mahāyāna Larger Sukhāvatī­ vyūha­ , seems to have been com- Buddhist literature, the word vyūha occurs very posed in (Buddhist) Sanskrit from the beginning. frequently as the final component of compounds The composer of the Smaller Sukhāvatīvyūha of proper names of buddhas and bodhisattvas, must have known that these two names referred e.g., Vimalakīrtinirdeśasūtra I § 4, Prabhāvyūha; to one and the same Buddha, but felt more com- I § 4, Mahāvyūha; I § 4, Padmavyūha; III § 29, fortable using Amitāyus than Amitābha and might Śubhavyūha; VI § 13, 44a3, Ratnavyūha etc. This have tried to explain their being the same by means might have led Zhi Qian, the probable translator of the above-quoted sentences. of the Wúliàngqīngjìng Píngdĕngjué jīng, into My hypothesis of the development Amitābha > misinterpreting *Amitāhavyūha as being the Amitābhu > MI. *Amitāhu > *Amitā’u > Amitāyu Buddha’s full name. He might have thought that > Amitāyus20 was criticised harshly by Fujita as *Amitāha in the original text and Āmítuó 阿彌陀 “a solely linguistic, unacceptable assumption in Lokakṣema’s translation to be abbreviated forms with disregard for the development of the ideas of of *Amitāhavyūha. He then translated *Amitāha, the Buddha.”21 However, I cannot find grounds to while relating it to Skt. amita (“limitless”), as relate the change from “(limitless) light” to “(lim- wúliàng 無量. As he consistently interpreted vyūha itless) life” to any theory on the development of incorrectly in association with MI śuha (< śubha the ideas of the Buddha, except for his arbitrary, “bright”) and translated it in a peculiar fashion a priori one meant to be used to support his own as qīngjìng 清淨 (“pure”),27 he might have trans- Amitābha / Amitāyus theory.22 lated -vyūha in *Amitāhavyūha as such. Thus, he The original Indic form of the transliteration of changed āmítuó(fó) 阿彌陀(佛) wherever it occurred this Buddha, āmítuó 阿彌陀 (QYS. ʔâ mjie4[mjie:4] in Lokakṣema’s translation to wúliàngqīngjìng(fó) dâ)23 in Lokakṣema’s earliest Chinese transla- 無量清淨(佛) without exception.

123 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

II. On the Name of Amitābha’s land: dating from the middle of the 12th century are Sukhāvatī / *Suhāmatī entitled Amitābhavyūha-parivarta Sukhāvatī­­ vyūha, while the remaining 36, being paper The Da āmítuó jīng contains other notewor- manuscripts, which were copied between the thy transliterations which indicate that its end of the 17th century and the first half of the 34 original language was Middle Indic, most prob- 20th century, are entitled Amitābhavyūha- ably Gāndhārī, e.g., tíhéjiéluó 提惒竭羅 (QYS. parivarta Sukhāvatīvyūha, Amitābhasya vyūha- diei γwâ gjɐt[gjät3] lâ; *Dīvagara < Dīpaṃkara; parivarta Sukhāvatīvyūha, śrīAmitābhasya 300b21), hélóugèn 廅28樓亘 (QYS. ʔâp ləu Sukhā­ vatī­ -vyūha nāma mahāyānasūtra, śrī­ sjwän; *Avalo . . . svar < Avalokita-svara; madAmitābhasya tathāgatasya Sukhāvatīvyūha- 308b15, 21, 309a15) etc. Amitābha’s Buddha- mahāyānasūtra, Amitābhasya parivarta 35 land (Sukhāvatī) is transliterated as xūmótí 須 Su­khā­vatīvyūha-mahāyānasūtra. 摩題 (QYS. sju muâ diei; 303b18), from which A relatively old manuscript of a Tibetan one may assume the underlying form was translation­ of this scripture, preserved in the *Suhā˘ matī or °adī, where the intervocalic h is manuscript collection of the Tabo monastery silent, common in Gāndhārī. In the modified in the Spiti Valley, Himachal Pradesh, north- translation of the same text (T. 12, no. 361), Zhi ern India, bears Sanskrit and Tibetan titles, Qian changed this transliteration to xūmótí namely āryaAmitābhabyūha (i.e., °vyūha) 須摩提 (QYS. sju muâ diei; *Suhāmatī or °adī; nāma mahāyānasūtra and ’phags pa ’Od dpag 282c29, 288c9), while transliterating anew the myed kyi bkod pa zhes bya ba theg pa chen 36 name of the land as xūātí 須阿提 (QYS. sju ʔâ po’i mdo’. The Tibetan translation of this su- diei; *Suhā(v)atī, or °adī?; 288b25).29 The latter tra in the Kanjur is found in the fifth chapter of occurs only once in a verse, which is wanting the Mahāratnakūṭasūtra, under the titles ’Od in Lokakṣema’s translation.30 A later Chinese dpag med kyi bkod pa’i le’u (*Amitābhavyūha- translator, Nie Daozhen 聶道真 (fl. beginning of parivarta), de bzhin gshegs pa ’Od dpag med the 4th century c.e.), transliterated the name of kyi bkod pa’i le’u (*Amitābhasya tathāgatasya the same land as xūhēmótí 須呵摩提 (QYS. sju vyūha-parivarta), ’phags pa de bzhin gshegs pa xâ muâ diei; *Suhāmatī or °adī?; T. 14, no. 483, ’Od dpag (v.l. dpag tu) med pa’i bkod pa zhes 666c-1, 668a17) and xūhēmóchí 須訶摩持 (QYS. bya ba’i le’u (*ārya-Amitābhasya tathāgatasya sju xâ muâ ḍï; *Suhāmatī or °adī?; T. 24, no. vyūha nāma parivarta), de bzhin gshegs pa 1502, 1116b3).31 In Khotanese, we find the word ’Od dpag tu med pa’i sangs rgyas kyi zhing gi Suhāvatä32 for this Buddha-land. yon tan bkod pa (*Amitābhasya tathāgatasya From these facts, we can therefore presume buddhakṣetraguṇavyūha), de bzhin gshegs pa that Amitābha’s land had been called, at an ear- ’Od dpag tu med pa’i sangs rgyas kyi zhing gi lier stage, *Suhāmatī (or °adī) or *Suhāvatī (or bkod pa’i le’u (*Amitābhasya tathāgatasya °adī), where the h is silent, but not Sukhāvatī. buddhakṣetravyūha-parivarta). Although Zhi Qian translated it as ānlèguó 安樂 It should be noted that in the Tibetan transla- 國 “the Land of Happiness” (288c6) in one verse, tions, this sutra is not entitled Sukhāvatīvyūha, we are not sure whether the name was originally but is referred to as Amitābhavyūha or Amitā­ meant as such or not.33 bhasya vyūha. Also, in the Sanskrit palm-leaf manuscripts and some of the paper copies, it is entitled Amitābhavyūha-parivarta Sukhāvatī­ vyūha and Amitābhasya vyūha-parivarta Su­ III. On the Titles of the Chinese khāvatīvyūha. From these facts, I presume that Translations of the Sukhāvatīvyūha this sutra originally had the title Amitābhavyūha, Amitābhasya vyūha or the like37 and the name Though the Sanskrit version of the Larger Su­ Sukhāvatīvyūha, which appears only in the San- khāvatīvyūha is generally referred to as such in skrit manuscripts, was added later as its subtitle. modern books, the real titles found in the San- Moreover, the original texts of most of the Chi- skrit manuscripts and the Tibetan translations nese translations could have also been entitled vary greatly. Among the extant Sanskrit manu- Amitābhavyūha or the like. We shall now exam- scripts from Nepal, two palm-leaf manuscripts ine the titles of these translations.

124 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

III.1. Dà āmítuó jīng 大阿彌陀經 used at times in early Chinese translations as a (T. 12, no. 362)38 transliteration for Indic ap(a), av(a), was discon- tinued completely in later translations, leading This oldest Chinese translation by Lokakṣema39 it to be miswritten as gai 蓋 or gai 盖. Therefore is generally called Da āmítuó jīng 大阿彌陀經. it is quite probable that this not so often used However, the title Āmítuó jīng阿彌陀經 , as some character hé 廅 was miswritten as sa 薩, as their editions read, might have been the original trans- shapes resemble each other. Also, the character lation of the title of the underlying text. After gèn 亘, which in the first place had not been un- the appearance of the translation of the Smaller derstood properly, was then hyper-formed to tán Sukhāvatīvyūha by Kumārajīva, which was en- 檀, containing dan 旦 on the right side of the char- titled also Āmítuó jīng 阿彌陀經 (T. 12, no. 366), acter. This title, then, may mean “Amitā(b)ha the character da 大 “larger” was added to the title *samyāsaṃbuddha43 (and) Avalokitasvara save of no. 362 in order to differentiate it from no. human beings.” This was probably not the origi- 366, and thus the more familiar title Da āmítuó nal Indian title, but rather given by the translator jīng 大阿彌陀經 came into being. As stated above, or somebody else in a later period, summaris- from the title Āmítuó jīng 阿彌陀經, one may ing the content of the sutra—according to this presume *Amitāha-vyūha or *Amidāha-vyūha sutra (309a14f.), after Amitā(b)ha’s parinirvāṇa, (< Amitābhavyūha) as its underlying form. In Avalokitasvara succeeded him as a Buddha and the Koryo and Jin editions, there is a very long began to rescue (guòdù 過度) human beings and title at the beginning of the sutra, which reads other various sentient beings as Amitā(b)ha does. āmítuó sānyēsānfó sàlóu{fó}tán guòdù réndào jīng 阿彌陀三耶三佛薩樓{佛}檀過度人道經 (the second fó 佛 should be deleted according to an- III.2. Wúliàngqīngjìng píngdĕngjué cient catalogues), while in the other editions jīng 無量清淨平等覺經 (T. 12, no. 361) and also at other places (namely, at the end of the first and the beginning of the second juan This version is quite likely a modification as well as at the end of the sutra) in the same of the Dà āmítuó jīng 大阿彌陀經 undertaken Koryo and Jin editions, the title reads simply by Zhi Qian. As we have already seen above, Āmítuó jīng 阿彌陀經 or the Amitā(b)ha sutra wúliàngqīngjìng 無量清淨 was probably Zhi Qian’s instead. Also, an old catalogue, namely the Chu­ peculiar translation of *Amitāha-vyūha. The word sanzangji ji 出三藏記集, which was compiled in píngdĕngjué­ 平等覺 was the standard rendering the Liang Tianlan Period (502–519 c.e.), states: of samyaksaṃbuddha and so, from the Chinese “Āmítuó jīng 阿彌陀經. Two juans. The title in- title, we can reconstruct *samyaksaṃbuddhasya side the book reads āmítuó sānyēsānfó sàlóután Amitāhasya vyūha as its original form. However, guòdù réndào jīng 阿彌陀三耶三佛薩樓檀過度人 I assume that Zhi Qian simply added the word 道經,” 40 indicating that the long title was origi- píngdĕngjué 平等覺, based on sānyēsānfó 三耶三 nally a subtitle. Moreover, in the Pusa Shouzhāi 佛 (*samyāsaṃbuddha < samyaksaṃbuddha) in jīng 菩薩受齋經, which was allegedly translated the long title of the Da āmítuó jīng 大阿彌陀經. by Nie Daozhen 聶道真 (fl. beginning of the 4th century c.e.)—but is more likely a composition made in China ­under the influence of the Da III.3. Wúliàngshòu jīng 無量壽經 āmítuó jīng 大阿彌陀經 and others—we find the (T. 12, no. 360)

Ed ē or é following similar expression: “I take in Āmítuó­sānyēsānfótán 阿彌陀三耶三佛檀 in the This translation is most likely the work of Bud- west (and) the bo­dhisattvas, Hélóugèn 廅樓亘 and dhabhadra 佛陀跋陀羅 (359–429 c.e.) and Baoyun Móhēnabō 摩訶那鉢.”41 The original form of the 寶雲 and hence dating from 421 c.e. From the Chi- longer title and its meaning are both enigmatic. nese title, one might suppose *Amitāyur-vyūha However, I am inclined to agree with Prof. Chen as its original, but I presume the underlying title Jinhua’s hypothesis that sàlóu{fó}tán 薩樓{佛}檀 of this translation is more likely to have been is a corruption of hélóugèn 廅樓亘, which is an Amitābha-vyūha. The word wúliàngshòu 無量 incomplete transliteration of Avalokitasvara.42 壽 is found in many places in this translation, The character hé 廅 (QYS. ʔâp), which had been where the Sanskrit version reads Amitābha,44

125 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha which indicates the translators’ preference of SP(KN) = H. Kern and B. Nanjio, eds. Saddharmapuṇ­ wúliàngshòu 無量壽 over wúliàngguāng 無量光. ḍarīka. St. Petersburg, 1908–1912 Probably, owing to this preference, the title was Sukh(F) = Reading of the Larger Sukhāvatīvyūha in rendered as Wúliàngshòu jīng 無量壽經 instead of Fujita 2011 Larger *Wúliàngguāng jīng 無量光經. Sukh(SC) = Reading of the fragments of the Sukhāvatīvyūha possibly from the Bamiyan area in Harrison et al. 2002 T = J. Takakusu and K. Watanabe, eds. Taishō Shinshū III.4. Wúliàngshòurúlái huì 無量壽如來會 Daizōkyō 大正新修大藏經. 100 vols. Tokyo, (T. 11, no. 310.5) 1924–1934 Vimalakīrtinirdeśasūtra = Study Group on Buddhist This translation, belonging to the Chinese Ma­ Sanskrit Literature, ed. Bonzōkan Taisho Yuimagyō 梵藏漢対照『維摩經』 Vimalakīrtinirdeśa: Trans­ hāratnakūṭasūtra, was translated between 706– literated Sanskrit Text Collated with Tibetan and 713 c.e. by Bodhiruci (fl. 693–713). From this Chinese Translations. The Institute for Compre- Chinese title, one may presume *Amitāyuṣaḥ hensive Studies of , Taisho University. tathāgatasya vyūha-parivarta as its original title. Tokyo, 2004 However, just as in the case of the Wúliàngshòu ~ = Stem of a word, e.g., dharma~ jīng 無量壽經, the translator rendered Amitābha as ° = Except for letters, following or preceding the sign, wúliàngshòu 無量壽, because of its popularity in the word is the same as the preceding one China. I believe, therefore, that the underlying title α < β = The form α comes from β of this translation was probably *Amitābhasya tathāgatasya vyūha-parivarta, which corresponds to one of the above-quoted Tibetan titles. Notes * I am greatly indebted to Dr. Peter Skilling, Dr. Tim- othy Lenz, Ms. Liang Ye Tan and Rev. Peter Lait for III.5. Dàshèng wúliàngshòu zhuāngyán jīng carefully reading through this manuscript and making 大乘無量壽莊嚴經 (T. 12, no. 363) numerous valuable suggestions and corrections. 1. When a certain text was translated into Chinese This was translated in 991 c.e. by the Song-­ more than once and such translations are extant, we dynasty translator Faxian 法賢 or Dharmabhadra.­ are able to trace the development of that text as well. From this Chinese title, one may infer *Amitā­ For this purpose, it is particularly of great advantage yuṣaḥ vyūha-mahāyānasūtra as its original, that the Chinese translations usually supply the trans- though I presume that the underlying title of this lators’ names, which can be then dated, though these translation was probably *Amitābhasya vyūha- may not always be correct. mahāyānasūtra as well. 2. Cf. Karashima 1997: 138; 1999a: p. 141, n. 34. Cf. also Nattier 2006: 190ff. 3. Though this is attributed to Zhi Qian, it is most probably by Lokakṣema; cf. Okayama 1980, Kagawa Abbreviations and Symbols 1993: 17–29, Harrison 1998: 556–57 and Harrison et al. 2002: 179–81. The counter-arguments by Fujita (2007: Ap = M. E. Lilley, ed. Text Society. The Apadāna 39ff.), who clings to its credit to Zhi Qian, following of the Khuddaka Nikāya. 2 vols. London, 1925, the Chinese catalogues of the Chinese Tripiṭaka, are 1927 awkward. BHSG = F. Edgerton. Buddhist Hybrid Sanskrit Gram­ 4. E.g., “ Dharmākara, subsequently mar. New Haven, 1953 became a buddha, namely *Amitāha Buddha, who do. = ditto (possesses) the foremost wisdom, vigour and unparal- MI = Middle Indic leled light. The land, in which he lives, at present, is QYS = Reconstruction of the Qieyun 切韻 System. In extremely wonderful.” (其曇摩迦菩薩至其然後，自致得 this article, Qieyun System forms, reconstructed by 作佛，名阿彌陀佛。最尊智慧勇猛，光明無比。今現在所 Karlgren and revised by F. K. Li, are used. The fol- 居國土甚快善; 301a16f.). lowing further notational changes, made by Coblin 5. *Olokitasvara is an older form of Avalokitasvara, (1994), are also adopted here: 1. •- will be written which is, in its turn, an earlier form of Avalokiteśvara; as ʔ-, 2. ě will be written as e, 3. Division III and IV cf. Karashima 1999b. chongniu 重紐 finals will be redundantly identified 6. Some relatively new paper manuscripts have, in with superscripts “3” and “4” a verse, Amitābhasya (Sukh[F] 53.6), which does not

126 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha match the metre, while an older palm-leaf manuscript between 47–147 c.e. (cf. Falk 2011: 20–23; Falk and reads Amitāsya instead, which is a better and appar- Karashima 2012: 19–61; 2013: 97–169). This Gāndhārī ently more original reading (see Fujita 1992–1996, version is therefore probably contemporary with the II 990). In other verses, Amitaprabha, a synonym of original text of the Dàoxíng Bānrě jīng 道行般若經, the Amitābha is used: Sukh(F) 51.10 (Amitaprabhasya), oldest Chinese translation of the same scripture by 57.2 (do.). Lokakṣema in 179 c.e. Actually, the transliterations, 7. Sukh(F) 50.6 (Amitāyu), 10 (Amita-āyu), 51.4 found in the latter, suggest that its original language (Amitāyu), 8 (Amita-āyu), 53.8 (Amitāyu), 55.11 was Gāndhārī (cf. Karashima 2013). On the other hand, (Amitāyu). there are also fragments of a Sanskrit manuscript of the 8. Sukh(SC) 194.7 (prose: Amitābha~), 194.9 (do.), same scripture, discovered in Bāmiyān and now pre- 195.20 (do.), 195.24 (do.), 197.10 (do.); 209.9 (verse: served in the Schøyen Collection and elsewhere, which Amitāyu). are written in an old Brāhmī script of the Kuṣāṇa pe- 9. For example, the latter half of the following sen- riod, supposedly dating back to the second half of the tence: Sukh(F) 35.14–15. aparimitam eva tasya bhaga­ third century c.e. based on palaeographical evidence vata āyuṣpramāṇam aparyantam. tena sa tathāgato (Sander 2000: 288). Moreover, a Gāndhārī fragment ’mitāyur ity ucyate (“. . . the measure of the life span from Afghanistan, now in the Schøyen Collection, of that blessed one is unlimited. Therefore, that tatha- shows the Sanskritization of Gāndhārī, for example, gata is called Amitayus [‘Measureless Life’]” [Gómez the long vowel ā is marked by a point beneath the char- 1996: 83]), finds its parallel only in the Tibetan transla- acters. Also, the Sanskritic ligatures, such as śca, dhya, tion (cf. Kagawa 1984: 186–87). Just after this sentence, jña, ṣṭha, are used as well. Although a C14 test gave the form Amitāyus appears in the Sanskrit version its age between 72–245 c.e., its actual date is thought (Sukh[F] 35.17), while its Tibetan parallel reads ’Od to fall towards the upper end of this range, according dpag med (Amitābha) and the Chinese parallels read to its palaeographic features (Allon and Salomon 2006: āmítuó 阿彌陀, wúliàngqīngjìngfó 無量清淨佛 or sim- 289). From these, I presume that the Sanskritization of ply “the Buddha” (cf. Kagawa 1984: 188–89). Cf. also , which were originally transmitted in Fujita 1970: 307f. Middle Indic, including Gāndhārī, occurred around the 10. The form Amitāha is attested in Pali as a variant beginning of the third century c.e. reading for Amitābha, the name of a cakkavatti: Ap 13. In the earlier Chinese translations made in the 210.2 Amitābho (v.l. Amitāho) ti nāmena cakkavattī Han Dynasty, neither the transcription nor the transla- mahabbalo. tion of Amitāyus (“Limitless Life”) is found (cf. Nattier 11. In Middle Indic and Buddhist Hybrid Sanskrit, 2006: 196). This fact also demonstrates that Amitāyus when a short vowel is required for metrical reasons in is a secondary form of Amitābha. verses, a masculine nominative singular -aḥ / -o be- 14. As these verses are wanting in the Chinese comes -u or -a; cf. BHSG §§ 8.20, 8.22. translations, they are probably later interpolations in 12. According to Damsteegt, who investigated Hy- the Sanskrit versions. brid Sanskrit inscriptions from Mathurā, Buddhists 15. The abbreviations of the Sanskrit manuscripts there had already started using Sanskritized Middle In- of the Saddharmapuṇḍarīkasūtra are as follows: A2 = dic to write inscriptions between the latter half of the Ms. kept in the Asiatic Society, Calcutta, No. 4199; B first century and the second century and in the Kuṣāṇa = Or. 2204, Ms. kept in the British Library; Bj = Ms. period, i.e., 200–350 c.e. (He follows the theory that formerly kept in the Library of the Cultural Palace of Kaniṣka became the king of the Kushans in 200 c.e. On the Nationalities, Beijing (written in 1082 c.e.); C4, C5, the basis of Harry Falk’s research, however, Kaniṣka’s C6 = Mss. kept in the Cambridge University Library, reign is now believed to have begun in 127 c.e.) San- Add. No. 1683, No. 1684, No. 2197; D2 = Gilgit Ms. skrit came to be used in inscriptions. This practice of kept in the National Archives of India (New Delhi); K using Sanskritized language in Buddhist inscriptions = Ms. kept in the Tōyō Bunko, Tokyo (written in 1069/ spread from Mathurā to other regions. Parallel to this 70 c.e.; brought from Tibet by Rev. E. Kawaguchi); L1 = development, the language in Buddhist texts must Ms. kept in the Potala Palace, Lhasa; L2, L3 = Mss. now have been Sanskritized gradually in this period as well kept in the Norbulingka, Lhasa, (written in 1065 c.e. (Damsteegt 1978: 264–66). On the other hand, no San- and 1067 c.e., respectively); N1, N2 = Mss. kept in the skritization is found in the language of the northwest- National Archives of Nepal, Kathmandu, Nos. 4–21, ern inscriptions of the pre-Kushan Period (ibid.: 207–8). Nos. 3–678, respectively; O = the so-called Kashgar It is only from the age of the Kuṣāṇas that Sanskritized manuscript, actually discovered in Khādaliq but pur- Buddhist inscriptions started to appear (ibid.: 221). chased in Kashgar; P1 = Ms. kept in the Bibliothèque Recently, fragments of a Gāndhārī version of the Nationale, Paris, No. 138–39; R = Ms. kept in the Royal Aṣṭasahasrikā Prajñāpāramitā were discovered in Asiatic Society of Great Britain and Ireland, London, northern Pakistan and these are thought to date back, No. 6; T8 = Ms. kept in the Library of the University of with an 81.1% probability based on a C14 test, to Tokyo, No. 414.

127 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

16. Tib. Kanj. = The Dam pa’i chos padma dkar po 39. Cf. n. 3. in the Kanjur; see Karashima 2008: 215–16. 40. 《阿彌陀經》二巻 内題云《阿彌陀三耶三佛薩樓檀 17. Tib. Kho. = An old manuscript of a Tibetan 過度人道經》(T. 55, no. 2145, 6c25). translation of the Lotus Sutra from Khotan, now kept 41. 歸命西方阿彌陀三耶三佛檀，廅樓亘，摩訶那鉢菩 in the National Museum of Ethnography, Stockholm; 薩 (T. 24, no. 1502, 1116b28f.). I should like to thank see Karashima 2008: 215–16. Jan Nattier for drawing my attention to this reference 18. Fujita 2007: 287ff. in a personal communication. 19. Fujita 2007: 4, 140, 296. 42. Personal communication, May 2003. 20. Cf. Nattier 2006: 190. 43. Cf. Skt. samyaksaṃbuddha; Pāli sammāsaṃ­ 21. Fujita 2007: p. 247, n. 5. buddha; Gāndhārī samasabudha. 22. Fujita 2007: 249ff. 44. Cf. Fujita 1970: 301–2. 23. As Nattier points out (2006: 188ff; see esp. 194– 95), if the original word were Amita, as maintained by some scholars, Lokakṣema would have transliterated References it as *āmì 阿蜜 instead of āmítuó 阿彌陀 in the same way as āyì 阿逸 (Skt. Ajita) or bōluómì 波羅蜜 (Skt. Allon and M. Allon and R. Salomon. pāramitā). Salomon 2000 “Kharoṣṭhī Fragments of 24. Cf. Karashima 1997: 138; 1999a: p. 141, n. 34. a Gāndhārī Version of the 25. Nattier 2007: 382ff. Mahāpariṇirvāṇa-sūtra.” In 26. See III.1 in this article. Buddhist Manuscripts, vol. 1, 27. Cf. Nattier 2007: 371ff.; Karashima 2010: 18ff. ed. J. Braarvig et al., 243–73, 28. The Taishō Edition reads 蓋 instead, while its Manuscripts in the Schøyen ­basic text, the Koryō Edition (高麗藏), reads 盖. The Collection, vol. 1. Oslo. other editions correctly read 廅. Coblin 1994 W. S. Coblin. A Compendium 29. Xūātí 須阿提 might be a scribal error for xūhētí of Phonetics in Northwest *須呵題 (QYS. sju xâ diei; *Suhātī, *Suhādī), though no Chinese. Journal of Chinese extant editions or manuscripts support this reading. Linguistics Monograph Series, 30. Lokakṣema’s translation completely lacks vol. 7. Berkeley. verses, although they are found in the other versions, Damsteegt 1978 T. Damsteegt. Epigraphical including Zhi Qian’s translation. In other words, Zhi Hybrid Sanskrit: Its Rise, Qian, who otherwise just copied and modified the pre- Spread, Characteristics and existing Lokakṣema’s translation, translated the verses Relationship to Buddhist Hy­ to fill up his modified translation. brid Sanskrit. Leiden. 31. For other transliterations for *Suhāmati, found Emmerick 1968 r. E. Emmerick. The Book of in later Chinese translations, see Nishimura 1987: Zambasta: A Khotanese Poem 113ff. on Buddhism. London. 32. The Book of Zambasta §14.47 (Emmerick 1968: Falk 2011 H. Falk. “The ‘Split’ Collection 218). of Kharoṣṭhī Texts.” ARIRIAB 33. From these Middle Indic forms, we can suppose 14: 13–23 + pls. 7–8. *Sudhāvatī/*Sudhāmatī (< sudhā “the beverage of the Falk and H. Falk and S. Karashima. “A gods, nectar, ambrosia” + vat/mat suffix) as a possible Karashima 2012 First-Century Prajñāpāramitā alternative, though sudhā would have become *susa Manuscript from Gandhāra— in Gāndhārī. parivarta 1 (Text from the Split 34. Cf. Fujita 1992–1996: I vii–xii, III v–vi and Fujita Collection 1).” ARIRIAB 15: 2007: 19ff. 19–61 + pls. 5–7. 35. Cf. Fujita 1992–1996: II 1472–73, III 484. Falk and . “A First-Century 36. Cf. Harrison 2009: 94, § 1.3.3.5. There, byū tha Karashima 2013 Prajñāpāramitā Manuscript stands instead of byū ha (i.e., vyūha), though the for- from Gandhāra—parivarta 5 mer must be a misprint. (Text from the Split Collection 37. Cf. the Sanskrit title, ārya-Akṣobhyasya tathā­ 1).” ARIRIAB 16: 97–169 + pls. gatasya byūha nāma mahāyānasūtra and the Tibetan 52–53. one ’phags pa de bzhin gshegs pa Mi ’khrugs pa’i bkod Fujita 1970 K. Fujita 藤田宏達. Genshi Jōdo pa zhes bya ba theg pa chen po’i mdo’ found in the Ti- Shisō no Kenkyū 原始浄土思想 betan translation of the *Akṣobhyavyūha in the Tabo の研究 (A study of early Pure Collection; see Harrison 2009: 94, § 1.3.3.6. Land Buddhism). Tokyo. 38. I am in the process of publishing a Japanese an- Fujita 1992–1996 . The Larger notated translation of this text: Karashima 1999a–. Sukhāvatīvyūha: Romanized

128 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

Text of the Sanskrit Manu­ Daigaku Sōgōkenkyūjo Kiyō 佛 scripts from Nepal. 3 vols. 教大学総合研究所紀要 (Bulletin Tokyo. of the Research Institute of Fujita 2007 . Jōdo Sanbukyō no Bukkyo University) 6: 135–50; Kenkyū 浄土三部経の研究 (A 7: 95–104; 8: 133–46; 10: study of the Three Pure Land 27–34; 11: 77–96; 12: 5–20; 13: sūtras). Tokyo. 1–11; 14: 1–17; 17: 1–13. Fujita 2011 . The Larger and Karashima 1999b . “Hokekyō no Smaller Sukhāvatīvyūhasūtra. Bunkengakuteki Kenkyū (2)” Kyoto. 法華経の文献学的研究（二）–– Gómez 1996 L. O. Gómez. The Land of 観音Avalokitasvaraの語義解釈 Bliss: The Paradise of the (Philological remarks on the Buddha of Measureless Light. Lotus Sutra (2)—On the name Sanskrit and Chinese Versions Avalokitasvara). ARIRIAB 2: of the Sukhāvatīvyūha Sutras. 39–66. Honolulu. Karashima 2008 . “An Old Tibetan Harrison 1998 P. Harrison. “Women in the Translation of the Lotus Sutra Pure Land: Some Reflections from Khotan: The Romanised on the Textual Sources.” Jour­ Text Collated with the Kan- nal of Indian Philosophy 26.6: jur Version (4).” ARIRIAB 11: 553–72. 177–301 + 21 pls. Harrison 2009 . Tabo Studies III: A Karashima 2010 . “Amida Jōdo no Catalogue of the Manuscript Genhūkei” 阿弥陀浄土の Collection of Tabo Monastery. 原風景 (The original land- Vol. 1, Sutra Texts (Ser phyin, scape of Amitābha’s “Pure Phal chen, dKon brtsegs, mDo Land”). Bukkyū Daigaku sde, Myan ’das). Rome. Sōgōkenkyūjo Kiyō 佛教大学 Harrison et al. 2002 P. Harrison, J.-U. Hartmann 総合研究所紀要 (Bulletin of the and K. Matsuda. “Larger Research Institute of Bukkyo Sukhāvatīvyūhasūtra.” In University) 17: 15–44. Buddhist Manuscripts, vol. 2, Karashima 2013 . “Was the ed. J. Braarvig et al., 179–214. Aṣṭasāhasrikā Prajñāpāramitā Manuscripts in the Schøyen Compiled in Gandhāra in Collection, vol. 3. Oslo. Gāndhārī?” ARIRIAB 16: Kagawa 1984 T. Kagawa 香川孝雄. 171–88. Muryōjukyō no Shohon Taishō Nattier 2006 J. Nattier. “The Names of Kenkyū 無量壽經の諸本對照 Amitābha/Amitāyus in Early 研究 (A comparative study Chinese Buddhist Translations of the texts of the Larger (1).” ARIRIAB 9: 183–99. Sukhāvatīvyūha). Kyoto. Nattier 2007 . “The Names of Kagawa 1993 . Jodokyo no Seirit­ Amitabha/Amitayus in sushiteki Kenkyu 浄土教の成 Early Chinese Buddhist 立史的研究 [The origins and Translations (2).” ARIRIAB 10: development of Pure Land Bud- 359–94. dhism]. Tokyo. Nishimura 1987 m. Nishimura 西村実則. Karashima 1997 S. Karashima 辛嶋静志. “Dai “Gandāra go bukkyōken to Amida kyō ganmon yaku” kanyaku butten ガンダーラ語 『大阿弥陀経』願文訳 (A Japa- 仏教圏と漢訳仏典” (Gāndhārī nese translation of the vows and Chinese translations of in the Da Amituo jing). Kyōka the Buddhist texts). In Sankō Kenkyū 教化研究 117: 135–45. Bunka Kenkyūjo Nenpō 三康 Karashima 1999a– . “Daiamidakyō 文化研究所年報 (Annual Report Yakuchū” 『大阿彌陀經』訳 of the Sankō Research Institute 注 (An annotated Japanese for the Studies of Buddhism) translation of the earliest 20: 49–125. Chinese version of the Larger Okayama 1980 h. Okayama 丘山新. “Dai Sukhāvatīvyūha). In Bukkyō Amida kyō yakusha ni

129 karashima: On Amitābha, Amitāyu(s), Sukhāvatī and the Amitābhavyūha

­kansuru ichi kasetsu” 『大阿 Sander 2000 L. Sander. “A Brief Paleograph- 弥陀経』訳者に関する一仮説 (A ical Analysis of the Brāhmī hypothesis on the translator of Manuscripts in Volume I.” In the Da Amituo jing). Indogaku Buddhist Manuscripts, vol. 1, Bukkyōgaku Kenkyū 印度學佛 ed. J. Braarvig et al., 285–300. 敎學硏究 (Journal of Indian and Manuscripts in the Schøyen ) 28.2: 227–30. Collection, vol. 1. Oslo.

130